In typical user interactions with windows management and content retrieval, windows of information (e.g., a web page, a word-processing document, and a picture) are concurrently projected on a two-dimensional (2D) screen. These windows of information can be visualized and navigated via a “multiple layers” view, but even in the multiple layers view some of the windows are hidden by a top layer of windows. Unlike on a physical desk where documents and objects can be scattered around, in a conventional computer desktop user interface there is no fixed position given to inactive windows behind a top active window. Thus, the user cannot cognitively register background windows to a specific position in the physical space, making it difficult and inefficient for the user to choose among different windows.